TargetSpy: a supervised machine learning approach for microRNA target prediction

Abstract

BackgroundVirtually all currently available microRNA target site prediction algorithms require the presence of a (conserved) seed match to the 5' end of the microRNA. Recently however, it has been shown that this requirement might be too stringent, leading to a substantial number of missed target sites.ResultsWe developed TargetSpy, a novel computational approach for predicting target sites regardless of the presence of a seed match. It is based on machine learning and automatic feature selection using a wide spectrum of compositional, structural, and base pairing features covering current biological knowledge. Our model does not rely on evolutionary conservation, which allows the detection of species-specific interactions and makes TargetSpy suitable for analyzing unconserved genomic sequences.In order to allow for an unbiased comparison of TargetSpy to other methods, we classified all algorithms into three groups: I) no seed match requirement, II) seed match requirement, and III) conserved seed match requirement. TargetSpy predictions for classes II and III are generated by appropriate postfiltering. On a human dataset revealing fold-change in protein production for five selected microRNAs our method shows superior performance in all classes. In Drosophila melanogaster not only our class II and III predictions are on par with other algorithms, but notably the class I (no-seed) predictions are just marginally less accurate. We estimate that TargetSpy predicts between 26 and 112 functional target sites without a seed match per microRNA that are missed by all other currently available algorithms.ConclusionOnly a few algorithms can predict target sites without demanding a seed match and TargetSpy demonstrates a substantial improvement in prediction accuracy in that class. Furthermore, when conservation and the presence of a seed match are required, the performance is comparable with state-of-the-art algorithms. TargetSpy was trained on mouse and performs well in human and drosophila, suggesting that it may be applicable to a broad range of species. Moreover, we have demonstrated that the application of machine learning techniques in combination with upcoming deep sequencing data results in a powerful microRNA target site prediction tool http://www.targetspy.org.

Literature review of baseline information to support the risk assessment of RNAi‐based GM plants

“…The European Food Safety Authority reserves its rights, view and position as regards the issues addressed and the conclusions reached in the present document, without prejudice to the rights of the authors. (Xu et al, 2014a) Target Scan (Agarwal et al, 2015) TargetS py fail http://www.targetspy.org (Sturm et al, 2010) ToppMi R rank interactants as a function of their potential to impact a given biological systém yes http://toppmir.cchmc.org (Wu et al, 2014) Importantly, accurate bioinformatic prediction of miRNA-mediated repression is still problematic. This was shown, for example, during experiments with systematically generated artificial miRNAs targeting a desired gene .…”

miTALOS: Analyzing the tissue-specific regulation of signaling pathways by human and mouse microRNAs

“…In addition to the tissue filter, miTALOS also provides a pathway filter to restrict the functional analysis. miRNA target transcripts are obtained from five different prediction tools: TargetScanS , RNA22 (Farh et al 2005), PicTar (Krek et al 2005), PiTa (Kertesz et al 2007), and TargetSpy (Sturm et al 2010). Due to imperfect base pairing and the short length of binding sites, prediction of miRNA target genes often yields false positive target genes.…”

HTar: Hidden Markov Model Based MicroRNA Binding Site Prediction

“…Among the algorithms MiRanda, [10] TargetScan, [11] PicTar [6] and MTar [8] used a test for conserved regions as the initial screening step in the process of target prediction, whereas RNA22 [12] MicroTar [13] and TargetSpy [14] have considered factors other than conservation during this step. TargetScan, miRanda and PicTar perform an extensive search in the 3' UTR of mRNAs for probable targets.…”